Issues · www / Tesseract · GitLab

Perhaps use polyglot's language detection

#36 · created Dec 22, 2020 by Merlijn Wajer
- 1
updated Jan 05, 2021
Potentially use three letter language ISO639 codes for ocr_detect_lang

#35 · created Nov 19, 2020 by Merlijn Wajer
- 0
updated Nov 19, 2020
(Autonomous) Fraktur/script detection

#34 · created Nov 11, 2020 by Merlijn Wajer
- 0
updated Nov 11, 2020
Perhaps use character confidence somewhere

#33 · created Nov 09, 2020 by Merlijn Wajer
- 0
updated Nov 09, 2020
Set up tesseract repo clone and add a gitlab-ci.yml to automatically build the .deb

#29 · created Oct 31, 2020 by Merlijn Wajer
- 3
updated Jan 30, 2021
Fraktur testing collection

#24 · created Oct 24, 2020 by Merlijn Wajer
- 0
updated Oct 24, 2020
Consider disabling OpenMP alltogether for Tesseract

#23 · created Oct 24, 2020 by Merlijn Wajer
- 0
updated Oct 24, 2020
Consider setting `tessedit_do_invert=0` for books/microfilm without inverted text

#22 · created Oct 24, 2020 by Merlijn Wajer
- 0
updated Oct 24, 2020
Consider a testing version of tesseract (5.x) in a testing container

#20 · created Oct 22, 2020 by Merlijn Wajer
- 1
updated Jan 30, 2021
Consider writing PPI to hOCR files

#19 · created Oct 19, 2020 by Merlijn Wajer
- 0
updated Oct 19, 2020
Implement test phase on several languages to ensure that the modules work fine

#18 · created Oct 19, 2020 by Merlijn Wajer
- 0
updated Oct 19, 2020
Consider adding DPI info via page property 'scan_res'

#16 · created Oct 15, 2020 by Merlijn Wajer
- 0
updated Oct 15, 2020
(perhaps) Improve hocr files: add ocr-langs, ocr-scripts

#15 · created Oct 15, 2020 by Merlijn Wajer
- 0
updated Oct 15, 2020
Implement fallback procedure in case OCR fails

#11 · created Oct 14, 2020 by Merlijn Wajer
- 0
updated Oct 14, 2020
Implement/check page orientation

#3 · created Oct 14, 2020 by Merlijn Wajer
- 0
updated Oct 14, 2020
Document module features, limitations and input metadata/arguments in README.rst

#1 · created Oct 14, 2020 by Merlijn Wajer
- 0
updated Oct 14, 2020